Simplicial neural networks (SNNs), a generalization of graph neural networks to data that live on a class of topological spaces called simplicial complexes.
Firstly, thanks for your work. I have a few confusion regarding the paper and would like to clarify.
In the paper, the model seems to be tasked to impute data where some portion of the values were replaced with a constant.
However, i do not see how the data is imputed. This is because the paper introduced simplicial convolution which seemed to mean learning a kind of filter. Does imputing the right values just mean filtering the input p-cochains to have the 'right' value?
Another question is that, is there a reason why we impute data on the co-chains instead of chains?
Hi there. I've read the associated paper and am very interested in the methodology. In particular (for a course project) I would like to see if I can improve on your results (even in a small way) but am having trouble seeing how to run the given code on a test set. My questions are as follows:
Does the s2 dataset (for which processing instructions are listed in the README) correspond to the CC2 set from the paper?
Does your code have a quick way of reproducing your test metrics on the imputed data? I would be very happy to implement something like this.
Is there a straightforward path to partitioning the given dataset for train / test purposes? Or perhaps managing 2 datasets, one for train and one for test?
Any clarity here would be very much appreciated. Thanks for the neat paper!
I have been studying your paper recently, and I am amazed that it is so well done. The experimental results after convolution have been taken, but the results are saved in text format. I am curious about how to draw the accurate curve and the wrong curve in your paper, is it convenient for you to teach? Thank you very much!!
I find this work very interesting and I've been interesting in applying it to my context where graph networks are popular. for the 3 body interactions is it possible to effectively include "opening angles" or other 3 body interactions from the perspective of a certain node or do things need to be expressed as hypervolumes? What would be the best way to approach 3 dimensional euclidian spaces with this method?
I hope what I'm getting across makes sense! Again thanks for this work it seems like a natural extension of graph networks.